spateo.tdr.interpolations.interpolation_gp

Module Contents

Classes

Imputation_GPR

Functions

gp_interpolation(→ anndata.AnnData)

Learn a continuous mapping from space to gene expression pattern with the Gaussian Process method.

class spateo.tdr.interpolations.interpolation_gp.Imputation_GPR(source_adata: anndata.AnnData, target_points: numpy.ndarray | None = None, keys: str | list = None, spatial_key: str = 'spatial', layer: str = 'X', device: str = 'cpu', method: Literal[SVGP, ExactGP] = 'SVGP', batch_size: int = 1024, shuffle: bool = True, inducing_num: int = 512, normalize_spatial: bool = True)

normalize_coords(data: numpy.ndarray | torch.Tensor, given_normalize: bool = False)

inference(training_iter: int = 50)

interpolate(use_chunk: bool = False, chunk_num: int = 20)

spateo.tdr.interpolations.interpolation_gp.gp_interpolation(source_adata: anndata.AnnData, target_points: numpy.ndarray | None = None, keys: str | list = None, spatial_key: str = 'spatial', layer: str = 'X', training_iter: int = 50, device: str = 'cpu', method: Literal[SVGP, ExactGP] = 'SVGP', batch_size: int = 1024, shuffle: bool = True, inducing_num: int = 512) → anndata.AnnData

Learn a continuous mapping from space to gene expression pattern with the Gaussian Process method.

Parameters:

source_adata

AnnData object that contains spatial (numpy.ndarray) in the obsm attribute.

target_points

The spatial coordinates of new data point. If target_coords is None, generate new points based on grid_num.

keys

Gene list or info list in the obs attribute whose interpolate expression across space needs to learned.

spatial_key

The key in .obsm that corresponds to the spatial coordinate of each bucket.

layer

If 'X', uses .X, otherwise uses the representation given by .layers[layer].

training_iter

Max number of iterations for training.

device

Equipment used to run the program. You can also set the specified GPU for running. E.g.: '0'.

Returns:

an anndata object that has interpolated expression.

Return type:

interp_adata